Colorectal cancer will be diagnosed in over 129,000 Americans in 1999. To combat this disease, new avenues to decrease the risk of colorectal cancer, such as chemoprevention, are being explored by researchers. Non-steroidal anti-inflammatory drugs (NSAIDs) and hormone replacement therapy (HRT) have been shown to decrease incident colon cancer. Little is known of their effect on persons with a history of colon cancer which, fortunately, is a continually expanding population as survival has been significantly improving over the last twenty years. The objective of this epidemiologic study is to determine whether NSAIDs or HRT is associated with recurrence or survival among individuals diagnosed with colorectal cancer. The proposed research will establish a cohort of colorectal cancer patients treated with curative intent and create a comprehensive longitudinal database, including data on the ascertainment of subsequent adenomatous polyps, colorectal cancer and survival. The specific aims are: (1) to determine whether NSAID use deceases the risk of recurrence of colorectal cancer; (2) to determine whether HRT use deceases the risk of recurrence of colorectal cancer; (3) to determine whether NSAID use affects short-term survival; and (4) to determine whether HRT use affects short-term survival. The cohort will be established from colorectal cancer patients enrolled in two managed care organizations, Health Alliance Plan (Detroit, MI) and HealthPartners (Minneapolis, MN). Cohort subjects will be followed for at least five years for new evidence of disease, recurrence and survival outcome. Using automated pharmacy data, the timing of use and exposure to NSAIDs and HRT will be analyzed among cancer survivors, along with potentially confounding variables, in relation to these outcomes.